In 2022, as the AI boom was heating up and investors were throwing money at any startup with a chatbot, Craig Campbell made a contrarian choice. He walked away from what he calls a "blank check" from venture capitalists who wanted him to start another AI company. Instead, he launched a website — Past Maps — a niche service that lets users overlay historical maps onto modern-day ones. Two years later, that bet is paying off, proving that the old-school web is still alive and well in the right niches.
From Meta Engineer to Niche Publisher
Campbell is a former engineer at Meta and an experienced tech founder. He previously sold an e-commerce tool for Shopify businesses just as the AI frenzy was taking off. Instead of riding that wave, he chose to build something small, useful, and personal. "I had my prior VC investors breathing down my neck, going 'start something else. We'll write you a blank check,'" he recalls. But Campbell had his sights set on a different kind of project — one born from his hobby of metal detecting.
Metal detecting requires knowing where old structures, trails, and settlements once stood. Campbell built a tool for himself that would take publicly available historical maps from sources like the US Geological Survey and overlay them on a modern map. By adjusting the opacity, he could see exactly where a ghost town or an old road used to be, and then go search for artifacts. He shared his creation on Reddit with other metal detecting enthusiasts, and the response was immediate: other people wanted to use it too.
The Rise of Past Maps
Past Maps is deceptively simple. Users pick a location, then browse through a library of historical maps from different eras — 19th-century surveys, mid-20th-century topographical maps, and more. The tool overlays these onto a current satellite or road map, with an opacity slider to fade between past and present. The maps come from public data, but the interface and the infrastructure to serve them quickly are entirely Campbell's work.
The growth has been steady but impressive. In year one, Past Maps averaged 20,000 active users per month. By year three, that number had climbed to over 300,000. The income from subscriptions — $9 per week or $52 per year — is enough to support Campbell and his wife, who helps run the business. However, Campbell notes that he's earning about what he did as a mid-level engineer at Facebook (an E4 level). "I'm making the same as when I was like, an E4 at Facebook," he says. But for him, the trade-off is worth it: he's building something he loves and on his own terms.
How Organic Search Fueled the Growth
Past Maps' biggest source of traffic is Google Search. Campbell found early on that when people searched for historical information about a specific location — a church their grandmother attended, or abandoned mine sites in a particular county — Past Maps would appear high in the results. By carefully tagging his maps and webpages in a way that Google understands, he created a virtuous cycle. "As I started exploding out this data and making it finally available to Google and giving it a place on the web, traffic just started to build," he explains.
This approach is a throwback to an earlier era of the web, when sites grew by being genuinely useful and by optimizing for search engines without gaming the system. Campbell sees it that way: "This is how the web is supposed to work. This is actually the old school web. It is alive and well, but only in these really, really small niches." His site doesn't rely on display advertising, which has become dominated by Google's ad tech — something a U.S. court ruled an illegal monopoly in 2025. Instead, Past Maps is subscription-based, insulating Campbell from the volatility of ad markets.
Uses Beyond Metal Detecting
While metal detecting enthusiasts remain a core audience, Past Maps has attracted a wide range of users. Genealogy researchers use it to find the locations of ancestral homes. Environmentalists track changes in waterways — one user mapped the Duwamish River in Washington state, which was straightened for shipping, making it easier to understand how the landscape has been altered. Another daily user tracks old oil wells. Campbell says the variety of use cases has been surprising, but it all ties back to a fundamental human curiosity: how did the places we know look in the past?
For a casual visitor, the site is simply fun. I spent an hour exploring the history of my own neighborhood, watching 19th-century farms turn into suburbs. The ability to blend past and present at a swipe gives a tangible sense of time passing. It's a research tool, but it's also a toy — and that balance is why people keep coming back and paying for it.
Embracing AI, but on His Own Terms
Even though Campbell walked away from the AI gold rush, he's not a Luddite. He uses AI extensively to run Past Maps more efficiently. One example is customer service. He used to spend one or two hours every day personally answering every support email, writing lengthy messages rather than sending form responses. Now, a local agent model runs on his desktop, connected to his Gmail, and does front-line triage. It weeds out spam, identifies messages that need his attention, and drafts replies. Campbell reviews the drafts, hits send, and his daily customer service time has dropped to about 10 minutes.
He's also using AI to tackle a notoriously difficult problem: optical character recognition (OCR) for historical maps. "Cartographers are assholes," Campbell jokes, explaining that old maps have labels that curve along rivers, are cramped together, or use inconsistent spacing. Off-the-shelf OCR tools fail on them. He's had some success with modern large language models that use reasoning, but it's not a simple plug-and-play solution. "You have to still bring that human spark into the mix," he says. Campbell combines his own decades of experience with tools and experimentation to guide the AI toward better results. The human sensibility for experimentation — knowing when to try a different approach — remains critical.
The Recipe for a Modern Old-School Success
Campbell's story offers lessons for anyone trying to build a sustainable online business today. First, start with a passion project. He didn't set out to create a unicorn; he made something he needed for his own hobby. Second, make it genuinely useful for a specific audience. By solving a real problem for metal detecting enthusiasts, he found an initial community that spread the word. Third, optimize for search in a straightforward way — not through tricks, but by making data accessible and well-structured. Fourth, charge for value, not for attention. Subscriptions give stability and freedom from ad tech's ups and downs.
The old-school web may be under threat from AI summaries, Google's zero-click world, and the consolidation of traffic on a few platforms. But Campbell's experience shows that in small niches, the old rules still apply. A good idea, solid execution, and a willingness to do things differently can still build a business. He may not be making VC-level returns, but he's making a living doing something he loves — and that, for many people, is the real payoff.
Source: The Verge News